Your models are static. The world isn't.
Markets move every day. Models don't. Analysts still spend the morning figuring out which changes actually touch the book, opening models one by one, and reconstructing why the last forecast was there. IRIS connects the changing world to the forecasts, Methods, and history inside each model, so frontier AI can show what matters, where it matters, and why.
When the world changes, which of my models should change with it - and why?
That is not a request for a market recap. It is a portfolio question.
Companies report every 90 days. Rates, spreads, unemployment, commodities, and consumer data move in between. The analyst has to decide what is noise, what touches an estimate or thesis, and which workbook deserves attention before the day gets away.
The market move itself is not the answer. The question is where it intersects with the economic logic already expressed in the book.
A PM should not have to start the morning by opening twenty spreadsheets. IRIS brings the analytical work together: what moved, which names may be exposed, which Methods deserve another look, and where the existing view still holds.
A move in rates may matter to one company's financing cost and be irrelevant to another. The economic relationship - not the ticker on the screen - determines the work.
Each night, IRIS refreshes the external state, identifies which models are affected, and reruns the analysis that depends on what changed.
By morning, PM Search can ask across that refreshed analytical state: what changed, which models moved, which views held, and where attention belongs.
The PM sees where to look. The analyst sees why.
A generic frontier model can reason. But hand it a workbook cold and it first has to figure out the periods, formulas, forecast structure, and which parts of the file actually matter.
Some of its intelligence is spent rediscovering facts that software can establish exactly before it reaches the investment question.
IRIS does the model archaeology once. It prepares the problem before the model call, so the reasoning can be about funding cost, unit growth, margins, credit losses, pricing, volume, or operating leverage - whatever actually drives the forecast. We call this Semantic Compression. In tests it has led to a 20x reduction in token use.
The point is not a shorter prompt. It is a better-defined question and more of the model's intelligence spent on the economics.
Let computers establish the facts. Let frontier models reason about the economics.
Before an LLM reasons about the workbook, IRIS establishes what can be known exactly: periods, formulas, dependencies, forecast structure, historical relationships, and the relevant inputs and outputs.
The differentiated forecast comes from how the analyst weighs evidence, frames the drivers, and decides when the view should change. IRIS makes that approach explicit as a Method: readable, editable, and runnable through the workbook.
The PM View points to a name. The analyst opens the model with its structure already understood. The Research Rail sits beside the spreadsheet, showing how the forecast works, which Methods drive it, and what deserves a closer look.
IRIS can preview a proposed change before the analyst applies it to the model.
Why is the model different now? A changed estimate should not erase the reason the prior estimate existed.
Model Memory is how that thinking evolves.
Follow Versions from left to right. Open the Run beneath a change. Compare its Result with what came before and recover the reasoning that produced it. The lineage replaces overwritten cells and scattered notes with a record an analyst can actually revisit.
A move in credit spreads may matter to funding cost in one company, reinvestment yield in another, and expected losses somewhere else.
IRIS keeps those economic relationships with each model, making them durable and queryable across the book without assuming every company responds the same way.
That is what lets a PM ask one question across the book without flattening every company into the same macro sensitivity.
The portfolio becomes more than a collection of spreadsheets. Work done in one name becomes usable knowledge when the same driver matters elsewhere.
Understanding compounds across the investment process.
Excel stays the model.
Frontier AI provides the intelligence.
IRIS is the harness between them.
Less model archaeology. Faster detection of what matters. Fewer models opened unnecessarily. A durable record of why the view changed.